FORM - 2
THE PATENTS ACT, 1970 (39 of 1970)
PROVISIONAL SPECIFICATION (SECTION 10; rule 13)
"Non-Invasive Device For Estimation Of Hemoglobin Concentration Using 3 LEDs"


BIOSENSE TECHNOLOGIES PVT. LTD.,
A Company Incorporated under the Indian Companies Act,
D2, 303, Jaltarang, Lokpuram Gladys Alvares Marg,
Opp. Hiranandani Meadows, THANE (W)-400610,
Maharashtra State, India.

The following specification particularly describes the nature of this invention:


FIELD OF THE INVENTION
The present invention deals with the device related to non-invasive determination of hemoglobin concentration in blood using 3 LEDs and the method used therein. The present device relates to the non-invasive method of determining hemoglobin concentration in blood using the principles of reflectance spectrophotometry data of diffuse light of three wavelengths. The present invention uses a novel algorithm optimized for robust use in developing countries, utilizing matrix algebra and parametric evaluation techniques.
BACKGROUND OF THE INVENTION
Devices for non-invasive determination of the oxygen saturation of blood are generally known in the medical field. This device uses the difference of absorption between the two dominant species of hemoglobin viz. oxyhemoglobin and deoxyhemoglobin. A device, which can perform non-invasive measurement of hemoglobin saturation, is generally referred to as a pulse oximeter. It measures the oxygen saturation in blood and determines the pulse rate.
Spectrophotometric analysis is typically based on a model that assumes pure collimated light is reduced in intensity only by absorbing species. The intensity is reduced by an exponential process known as "Beer's law", wherein absorbance is proportional to concentration.
Photoplethysmography also forms the basis of obtaining signals in a non¬invasive way. Oxyhemoglobin absorbs more light at 940 nm and deoxyhemoglobin absorbs more light at 660 nm. This difference in absorption is utilized to obtain the oxygen saturation of blood. Tissue contains absorbing species other than hemoglobin too. However, it is possible to separate these using pulse plethysmography. It is done by getting the difference of absorption/reflectance from the peak of an arterial pulse and the

absorption/reflectance before an arterial pulse. This difference is attributed to the arterial blood at the site of measurement.
In photo-plethysmography reflectance or transmittance can be used to obtain a signal that is processed further to determine the amount of light that gets absorbed by the column of blood passing through the artery. Theoretically transmittance was ruled out and instead reflectance was chosen. The different layers of the finger collectively have a scattering effect on light and hence diffuse reflectance is the predominant phenomenon occurring in the human finger when light is made to pass through it.
The device under consideration uses the technique of photoplethysmography for obtaining signals to measure hemoglobin in blood. The technique involves the use of 3 LEDs to illuminate a portion of the finger. Human skin is highly variable media, involving multiple scatterers like melanin etc. The light emitted by the LEDs is either scattered or absorbed (considering negligible luminescence as it passes through the anatomical mediums.) Only a portion of the incident light passes through or reflects back from the finger that is measured by a photo-diode either in the form of transmitted light or as a component in the form of reflected light.
Generally finger probe is used with above-mentioned criteria in mind. Yet inaccurate results are obtained if the probe is loosely held against the finger and due to physiological variations in the finger. Variation in light of the surrounding region can also lead to inaccurate results. Variation also occurs because of the pulsating nature of blood passing through that region. The volume of blood peaks/rises with the systolic and drops to a minimum at the diastole. The absorption of light increases with an increase in the volume of blood passing through this region. A waveform is obtained because of the varying absorption of light with each pulse. This waveform helps in establishing the component of light getting absorbed by the pulsating volume of blood and further in establishing the concentration of specific components in blood. However, there are some

constraints that need to be looked into. Major reasons for inaccurate results in determining hemoglobin levels are: a) variability in skin color due to the dissimilarity in melanin concentration and b) variation in finger size across population.
PRIOR ART
There are various inventions made in this field previously but the present invention is novel with respect to the prior inventions as described below.
U.S. Patent no. 4819752 discloses a pulse oximeter type machine which measures hemoglobin oxygen saturation using these principles. The present invention differs from the above said patent in that it measures the total hemoglobin value (which is an absolute value, where as oxygen saturation is a ratio). Also, the present invention uses a diffuse light of different wavelengths.
Similarly, U.S. Patent no. 4805623 discloses a spectrophotometric method of measuring the concentration of a dilute component in a light or other radiation scattering environment. The present invention differs in a way that it does not use simultaneous measurement of dilute and reference components .
Also, U.S. Patent no. 5377674 relates to a method of non-invasive and in-vitro hemoglobin estimation. This patent differs fundamentally from this invention in that it uses second derivatives of absorbances of lights of wavelengths of 800-900 nm and 1300-1700 nm. Differences arising due to variations in carboxyhemoglobin / methemoglobin levels are not factored in.

DISADVANTAGES OF PRIOR ART
The present invention prefers the use of reflectance spectrophotometry over transmittance due to the highly variable composition of the skin, and concentration of scatterers like melanin and other chromophores.
The finger probe is the interfacing unit that determines the value of hemoglobin in blood. An accurate result can only be achieved if the design of the probe meets some of the following physical constraints:-
1) The probe has to be used across population and there is a lot of variation in terms of finger size, color, skin thickness and even physiology.
2) The probe has to fit snugly to the patient's finger so that there is no relative movement between the finger and the probe.
3) Ambient light interferes with the light emitted by the LED's. As a result the light captured by the Photodiode contains not only a component of unabsorbed light as emitted by the LED's but also a component of unabsorbed ambient light leading to inaccurate results.
DESCRIPTION OF THE INVENTION
Reflectance spectrophotometry uses the scattering and backscattering of photons to give a quasi absorption spectra of the blood. Blood is essentially a biochemical matrix, if the DC signals are effectively filtered off then the concentration of hemoglobin can be estimated with a level of accuracy using wavelengths from 650-1000 nm in order to maximize the accuracy of the device.
The proposed invention uses the phenomenon of reflectance to measure the amount of light that gets absorbed by the components of blood. This is because there is considerable back scattering involved as the light traverses the finger. The amount of light that gets absorbed by the non-pulsatile organic tissue of the

finger remains the same irrespective of the amount of blood passing through the zone of the finger under observation.
A new probe design has been introduced that uses three LEDs instead. Each LED chosen is of a different but specific wavelength. The wavelengths chosen correspond to the isosbestic point of the different hemoglobin types. The LED's along with the photo-diode fulfill multiple criteria. Each LED illuminates a common region of the finger such that the light received by the photodiode during successive pulses and from each LED is from a particular segment of the artery. The inclination of each of the LEDs is determined by the Monte - Carlo simulations. The three LED's are inclined such that each of them is equidistant from the photodiode as well as each other. The inclination of the center LED and the photodiode is designed so that the line of incidence and the line of reflectance intersect at a height equal to the assumed average depth of the artery and at the point that lies directly above the mid point of the distance between the LED and the Photodiode. The side LEDs points towards this projected point of intersection of the middle LED and the photo-diode.
The assumption of artery depth is sufficient criteria for evaluating because light undergoes heavy diffusion the moment it enters the finger. Equal penetration of light for the 3 LEDs becomes more crucial till the projected intersection point, which is calibrated by the receiving signal for every pulse cycle and by varying the resistance that allows for equating the penetration of light emitted by LED of different wavelengths.
Ambient light produces noise with the signal obtained at the Photo-diode. This noise is in the form of direct noise that the photodiode receives when light enters the space between the photodiode and the skin. Indirect noise is when light penetrate through the finger, diffuses and accompanies the reflected light of the LED's to form the signal obtained at the photodiode.

Indirect noise is prevented by enveloping the finger with a black opaque sheath that does not allow the ambient light to penetrate as well as reduces the scattering of ambient light around the LED and probe cluster.
Direct noise is prevented by pressing the LEDs and photodiode against the skin so that there is no gap between the two. A secondary barrier is introduced that acts like a fortress wall. This wall runs along an elliptical periphery, encloses the LEDs and the photo-diode within and presses against the finger. A diamond and a rectangular periphery were given away because the elliptical form fits the finger better irrespective of finger size.
A base supports the diodes into position. The base together with the walls encloses the diodes and when pressed against the finger forms an isolated space where no ambient light can enter. This isolated space is what is coined as the optical chamber. The uniqueness of the optical chamber lies in the fact that it prevents all ambient light in the form of noise to enter the photodiode. Direct noise from the ambient light is further prevented from entering the photodiode by enclosing the chamber in a space between the finger and the probe body
In the present invention the probe body is made out of a flexible and elastic material in the form of a finger glove. This finger glove has an integrated strap that can be adjusted to accommodate variations in finger size, thus making it a unique design to fit all finger sizes. The elasticity in the material ensures that the probe can be stretched and fitted snugly over a finger. It not only ensures that the relative movement between the probe and the finger is nullified but also that there are minimal chances of ambient light escaping through the space between the finger and the probe. The present invention uses three LEDs that cancel out the effective path length traveled by the photons in order to overcome the barriers existing in the present models.

To,
The Controller of Patents,
The Patent Office, Mumbai.